Is mortals' chiefest enemy.
William Shakespeare, Macbeth.[1]
|
| And you all know security Is mortals' chiefest enemy. William Shakespeare, Macbeth.[1] |
In this chapter, and in the two that follow it, empirical findings that mainly concern the first condition will be discussed. The second stipulation will be dealt with in Chapter 11. The traditional policy toward improvement of road safety--although the denominator of safety has rarely been clearly identified--goes under the common label of the "Triple E" approach: Engineering, Education and Enforcement. Note that in this approach there is no specific reference to the concept of motivation--the concept that, according to risk homeostasis theory (RHT), is the most relevant to safety of all.
As a further case in point, the work of Austrian researcher Lieselotte Schmidt[3] is explicitly aimed at a reduction of the target level of risk, not by means of material inducements (such as monetary rewards, merchandise or extra holidays[4]), but through the enhancement of moral judgement, safety-consciousness, awareness of personal responsibility, and consideration for other people as well as for the ecological environment.
There can be no question that Schmidt's message is much more palatable to the cultured mind, at both the emitting and receiving end of the education process, than "materialistic bribery into safe conduct" and rewarding people for virtues they are morally obliged to display of their own accord. But there remains the practical question as to how effective this approach can be in a human condition in which stimuli generally tend to score a greater effect as they hit the subject closer to the wallet.
In a more optimistic vein, however, it is interesting to note that behaviours conducive to economic success, or reflecting it, often become "the proper thing to do." A deep tan and a slender body may be fashionable in one society while corpulence and an alabaster skin is fashionable in another. Wearing a white shirt, having long and polished fingernails, or walking on high heels may elicit social admiration while few of the admirers consciously realize that the attractiveness derives from economic aspiration or success.
In northern Europe or Canada ("the Scandinavia of the Americas"), to be tanned in winter-time is a signal of being prosperous enough to spend a vacation on the ski slopes or in the south; in warmer climes, to be tanned marks a person as someone who is demeaned by having to make a living labouring the land. It will be interesting to see how the increased threat of skin cancer is going to affect people's admiration for a "beautiful" tan, because it may come to signal lack of care about oneself rather than an indicator of success. An alabaster skin may gain in physical attraction. "She had so transparent a skin that I almost could have felt her pulse with my eyes, I believe."[5] Already gone is the image of a female smoker as "a sophisticated lady", smoking now being viewed as self-destructive. The turn-on has turned into a turn-off.
The extension of material rewards to people who have no accidents may eventually lead to a situation in which cautious conduct will be seen as "the proper thing to do". To play it safe will less often be viewed as sissy and more often as sensible. Thus, the value-oriented educational approach taken by Schmidt and the material-incentive approach advocated by this author may ultimately lead to similar end results. Which of the approaches is the more (cost-) effective is a matter for empirical investigation.
This, however, is only one of the consequences. Suppose that, as a result of training the risk-underestimators, the accident rate in the population as a whole drops to a level that is noticeably lower than before. People would then experience a reduced need for caution, and the per capita accident rate would not change in the long run.
There is another consequence. It should not be overlooked that there are other people who overestimate the danger of particular conditions or manoeuvres, and correction of risk perception on the part of the risk-overestimators would be expected to lead to an increase in their accident likelihood. "He who fears dangers will not perish by them," according to Leonardo da Vinci. People with a strong fear of flying are unlikely to become the victims of airplane crashes. Those who are afraid of heights will not climb a tall ladder and thus run no risk of falling from it. Agoraphobics rarely become pedestrian casualties. A successfully treated agoraphobic runs a greater likelihood of becoming a pedestrian casualty than when terrorized by this phobia. "Just as courage imperils life, fear protects it",[7] but Leonardo would probably agree that excessive fear also hampers survival: in order to maximize potential benefit, people must optimize the amount of risk they take. As optimal risk is greater than zero risk, the occurrence of accidents is an inevitable consequence of the struggle for success. The challenge is to optimize accident risk, not to minimize it. Nothing ventured, nothing gained.
So, when people whose overestimation of risk has been successfully corrected take to the roads again, their accident frequency should rise. If the aggregate accident rate rises enough to be noticeable to people in general, then people in general will become more cautious and the aggregate accident rate will be expected to return to the old equilibrium between target risk and experienced risk.
It would seem, then, that little lasting benefit for safety may be expected from training. In the jargon of control theory, this is because the skill factors are outside "the closed loop" (see Figures 2.1 and 4.1). Perceptual, decisional and motor control skills can only produce fluctuations in the system output--as graphed in Figure 2.2--but no lasting and stable change.
Nonetheless, driver training is often viewed as an accident antidote. Some countries even demand formal driver training by law--as a prequisite for taking a driver's test. That the wisdom of this is questionable is illustrated by the 1983 Québec law that made driver training courses mandatory for anybody wishing to obtain a driver's licence. Prior to that date, this was required only of 16- and 17-year-olds.
The effect of the change in legislation was investigated by researchers at the Université de Montréal, who concluded that the new requirement had no appreciable effect on the frequency or severity of accidents amongst newly licensed drivers who were 18 or older. That was one major finding, but there was another. The risk of accident actually increased for 16- and 17-year-olds. This was attributed to the fact that the new legislation resulted in an increase in the number of young people obtaining a licence before age 18. The economic incentive for waiting until that age--and avoiding the added cost of taking training at a registered driving school--was no longer in existence.[8]
Other countries leave it to prospective drivers to decide how they wish to be trained: formal driver training offered by high schools, by private driving schools, or informal instruction from one's parents or an older sibling. In some countries, insurance premium discounts are offered to novice drivers who have undergone driver training in high school. This may be viewed as a reflection of the belief that formal training is beneficial to safety, or it may simply be a commercial tactic to enlarge the market and sell more insurance. Belief in the efficacy of driver training would seem to be the reason for governments and/or school boards to subsidize driver training in high school. An alternative reason might be to enhance the mobility of youth.
So, there are two opposing views, the prevailing one maintaining that driver training is productive towards safety, and the view held by RHT that it is not. Further empirical facts should be interesting. A British study compared the accident experience of drivers who had been trained in one of the following ways: (a) driving school only, (b) with friend or relative only, and (c) combined tuition.[9] Accident experience was expressed in terms of the average number of km driven per accident occurrence, so the higher the rate, the safer. In the first group (a) the average was 19,392 km; with friend or relative (b) it was 22,801 and for combined training (c) 14,536 km. In other words, the safest performance was found in group (b)--amongst those novice drivers who had not obtained any professional instruction.
In Ontario, self-reported and officially recorded accidents were compared between (a) some 800 motorcyclists who had graduated from the Motorcycle Training Programme conducted by the Ontario Safety League, and (b) some 1100 motorcyclists who had informal training only. There was no difference in accident experience between the two groups.[10]
Although both of these studies fail to support the notion that formal training makes people more accident-free, it may countered that we are dealing with self-selected samples. People chose one form of training or another, and it is conceivable that this choice was made on the basis of personal characteristics that are associated with accident involvement.
Therefore, we turn to a real-life experiment which was conducted in Georgia, the Georgia with Atlanta as the capital, not Tbilisi. In DeKalb County in this state, trainee drivers were not allowed to choose, but were arbitrarily assigned to one of three training conditions. The first was called the Safe Performance Curriculum [sic]. This had been developed by the National Highway Traffic Safety Administration and was considered the most advanced and thorough driver training programme available in the USA. It involved about 32 hours of classroom instruction, 16 hours of training on a simulator, 16 hours of instruction on a driving range, 3 hours of practising emergency manoeuvres, and over 3 hours of driving on public roads, including 20 minutes of night-time driving.
The second was a minimal training programme for providing the skills that are necessary to pass the driver's test. Since this included no more than a total of about 20 hours instruction in the classroom, on the driving range and on the simulator combined, and only one hour of actual driving by the student, this course contained considerably fewer hours of instruction than the typical high school training course.
A third group of students received no formal training and were expected to be taught by their parents (although some may have taken training in private driver training schools).The three groups were matched in terms of gender, age, grade point average and parents' socio-economic status.
Each of the groups in this four-year study consisted of about 5500 students. Learner-drivers who had graduated from the special training programme obtained their driver's licences sooner and had significantly more crashes than those who had received minimal training or no high school driver training at all. There was no significant difference in the crash involvement of the latter two groups.
It is obvious that the results of this large-scale field experiment do not support the notion that improved driver education helps prevent accidents. In passing, we may note that the same holds for post-licensure defensive driving courses.[11] This is in agreement with what one would expect on the basis of RHT. But why was the safety record of the graduates of the supposedly superior training course actually worse? According to some analysts of the data, a "possible explanation is that, compared to students receiving the [intensified course] or even more typical high school driver education courses, students in the [minimal] course may have finished the course with less confidence in their driving skills because of their limited behind-the-wheel instruction; this may have resulted in a slower rate of licensure and more caution during their initial periods of solo driving".[12]
This would seem a plausible interpretation, because on a separate test of driving skill, the group with the intensive training performed better than the group with minimal training, and the latter did better than the group without any high school driver training. So skill could not have made the difference. Most likely, the difference was due to overconfidence, or underestimation of risk due to overestimation of one's own ability, in this case, inspired by having had the privilege of receiving "superior" instruction. Alternatively, as it is difficult to accept that training could make people worse, one could reason as follows. Suppose that beginner drivers as a group more often overestimate accident risk than underestimate it. All else equal, they will thus incur fewer accidents than agrees with their target level. Now assume that one of the effects of training is to correct this overestimation of risk. As a result, additional training should led to an increase in accident rate precisely because the graduates have become more competent, more competent risk takers that is (see Figure 4.2). Better driving skill--risk perception included--does not necessarily mean fewer accidents.
Other studies, too, show that better driving skill is not associated with greater safety. The number of different types of faults made by drivers during the licensing test in Britain were unrelated to their subsequent accident record, but it should be noted that their faults were not serious enough to make the candidates fail the test.[13] A sample of over 22,000 drivers in Illinois showed the same lack of relationship between mistakes made on the written licensing test and subsequent frequency and type of moving violations over a four-year period.[14] Similarly, being at fault in an accident was not more common in drivers with less than average knowledge of the written driving test.
A study of the accident record of racing drivers in the USA found these drivers to be involved in crashes considerably more often than drivers on average, matched for age and gender. We are referring here, of course, to the accidents of licensed racing drivers while driving public roads, not race tracks.[15] They were found less safe per km driven as well as per person-year.
In our view, the increased accident frequency of the racing drivers is not due to their superior driving skill--since accident frequency in RHT is regarded as ultimately independent of skill--but can more likely be attributed to a greater-than-average acceptance of risk, which induced them to pick up the activity of car racing to begin with. At their level of skill, driving like the average driver may be intolerably boring. Imagine you can master the piano like a Beethoven and all you are allowed to play is "Twinkle, twinkle, little star"! It is not surprising that the racing drivers incurred many more traffic fines than drivers in general, especially for speeding.
That it often is not, and that the sense of safety may be greater than actual safety, has not only been observed for traffic, but also for the home medicine cabinet, the use of condoms, and for other equipment designed to contain danger. It is this situation that can be fatal.
Suppose that yield signs at intersections are replaced by stop signs, that traffic lights are installed at crossings where relatively little traffic passes, that curved roads are straightened out, that slippery road sections are replaced by high-friction road surfaces, that unmarked pedestrian crosswalks are equipped with zebra stripes, and the like. According to risk homeostasis theory, such modifications of the road environment should not improve the accident rate per person, nor should they make the it any worse. Nonetheless, there have been cases in which such changes have been reported to lead to increases in accident rates at the locations concerned. According to some of the proposed explanations: "A straight, monotonous road lulls the driver"; marked crosswalks give pedestrians "a false sense of security that the motorist can, and will, stop in all cases".[16]
Such changes in accident rates are only compatible with RHT provided they are local, rather than general throughout the entire road network, and if they are modest and/or short-lived. In other words: provided the difference in accident rate is less than the "just noticeable difference" (see Sections 2.2 and 4.3).
Other victims of the "lulling effect" have been reported, e.g. children under the age of five. In 1972, the Food and Drug Administration in the USA ordered manufacturers of painkillers and other selected drugs to equip their bottles with "child-proof" lids. These are difficult to open for children (and sometimes for adults as well) and often go under the name of "safety caps," a misleading name, as we will see. Their introduction was followed by a substantial increase in the per capita rate of fatal accidental poisonings in children. It was concluded that the impact of the regulation was counterproductive, "leading to 3,500 additional (fatal plus non-fatal) poisonings of children under age 5 annually from analgesics".[17] These findings were explained as the result of parents becoming less careful in the handling and storing of the "safer" bottles". "It is clear that individual actions are an important component of the accident-generating process. Failure to take such behavior into account will result in regulations that may not have the intended impact". Indeed, safety is in people, or else it is nowhere.
If parents can be blamed for the lack of effectiveness of safety caps, does a government that passes such near-sighted safety legislation go guilt-free? Does an educational agency that instills a feeling of overconfidence in learner drivers go guilt-free? Does a traffic engineering department that gives pedestrians a false sense of safety remain blameless; or a government that requires driver education at a registered driving school before one is allowed to take the licensing test? Is it responsible to call a seatbelt a "safety belt", to propagate through the media such slogans as "seatbelts save lives", "speed kills", "to be sober is to be safe", "use condoms for safe sex", or others of the same ilk?
In any event, it is interesting to note that accident countermeasures sometimes may increase danger, rather than diminish it. If stop signs are installed at junctions in residential areas and at all railway crossings that have no other protection, if flashing lights appear at numerous intersections, if warning labels are attached to the majority of consumer products, these measures will eventually lose their salience and their credibility. They amount to crying wolf when no such beast is in the area. And in the rare event it is, the warning will no longer be received and there may be a victim.
This is why over-use of warnings may be dangerous. A warning that is not perceived as needed will not be heeded--even when it is needed. "A warning can only diminish danger as long as there is danger." This is the paradox of warning. It sounds puzzling, but what it means is that warning signs can only make people behave more cautiously if they agree that their behaviour would probably have been more risky if they had not seen the warning sign. Similarly, "a warning can increase danger when it overstates danger", meaning that a person's behaviour may become less cautious if that person has learned that the danger is usually less great than stated in the warning. The somewhat puzzling element in these statements is due to the word "danger" having been used with two different meanings. The first meaning is a "lack of caution in the face of an external threat, increasing the likelihood of accident", while the second use of the word refers to some "external threat".
In some cases, a warning sign will have no effect on the frequency or size of the external threat. A stop sign placed on the roadway at a railway crossing will not normally affect the number or speed of approaching trains. Consider, however, a street crossing in a residential neighbourhood without any special "protection". The basic rule applies: traffic coming from the right has the right-of-way (or from the left, in some parts of the world). The city engineering department installs a yield sign on one of the two streets. Suppose you are approaching the junction while driving on the street with the yield sign. The chances of colliding with a car at a ninety-degree angle are now increased if you merrily proceed through the crossing without appropriate checking, because cross-traffic always has the right-of-way, regardless of whether it comes from the right or the left, and is less likely to be watching out for you.
For the cross-traffic the situation is different. The threat of angular collision while proceeding through the intersection without appropriate checking is diminished due to the protection of the yield sign. So, the new situation does not call for the same level of caution; your cross-traffic can afford to reduce their attention or move faster and still proceed at the same risk as before the installation of the yield sign. As a result, if anybody driving your street drives past the yield sign into the crossing while cross-traffic is present, a collision is more likely to happen. Not surprisingly, then, the installation of yield signs has not been found to reduce accidents at intersections.[18] A yield sign is useless from a safety point of view, at least in the longer run, while it may have the merit of improving traffic flow.
And so we come to the conclusion that warning signs have little or no lasting effect upon safety, regardless of whether their installation alters the external threat. Maybe they are more effective in serving a rather different purpose, namely, that of reducing the legal liability of a city's engineering department, or of a product manufacturer, if an accident happens. In the case of litigation, they can point out: "Yes, at that intersection accidents are possible, but we told you so." Seeking protection from legal liability and seeking to protect the public from accidents should not be confused. Clearly, the two purposes demand rather different action.
In the next chapter we will discuss some technological countermeasures that, despite great expectations on the part of those who developed them, did not seem to have a measurable effect on safety, even in the short term. First, however, we will deal with the question of the effects of mass media communications on health and safety habits.
In contrast to engineering advances in mass communications technology, social-science information on how people respond to mass-messages has been slow to develop. There is, as yet, no firm and detailed body of knowledge telling us what beneficial or detrimental effects, if any, may be expected from a given mass media communication. While the citizenry is continually exposed to print and broadcast messages, the consequences are mostly unknown.
It is true that large numbers of commercial advertising messages are tested for their effects on consumers, but the resulting knowledge may not be of much use to society as a whole because this knowledge remains largely confidential. At the same time, many public service messages for the promotion of health and safety remain untested and their effects are unknown. Some of the safety and health publicity may have no effect, or may even lead to attitude or behaviour change contrary to that desired. Without programme evaluation, this can never be determined.
Because of the limited pertinence of intermediate criteria such as those listed above (with the exception of k), these are all of a remote, intermediate, surrogate, or proxy nature. Data from empirical studies on the effects of mass-media messages should be evaluated with this in mind.
There are various factors that are responsible for the reduced pertinence of the intermediate yardsticks in common use. The relationship between attitude and manifest behaviour is often weak. One factor responsible for this may be the particular measurement procedures used. Another is that behaviour is also being affected by influences other than attitude, such as the person's perceptions of the norms held by others and the social pressure thus produced.[22]
Attitude change may lead to behaviour change, and vice versa as well. Studies show that if one is made to commit oneself to some behaviour because of a small inducement, one's attitude towards that behaviour may change. Some authors, therefore, describe attitude change as an iterative process: a communication brings about a minor change in attitude towards an advertised product or recommended practice that is enough to make the recipient willing to try it out.[23] The experience that follows with the product or practice causes the person to modify or solidify the original attitude, and this leads to additional experience with its effects on attitude, and so on:
Other yardsticks of effect seem even more questionable. Ratings of message effectiveness made by experts don't necessarily predict the behaviour change the messages produce, even when the effectiveness ratings are pooled across a panel of safety experts.[26] This is why, when it comes to evaluating the persuasive effect of messages, there is no truly dependable alternative to empirical testing of the behavioural effects on the recipients.
Finally, it is important to stress that a behavioural effect does not necessarily imply a change in the ultimate criterion k. This is because lifestyle-related disease, injury or mortality does not exclusively depend upon specific behaviours. Smoking is not the only habit leading to premature lifestyle-dependent death. Seatbelt use or sobriety do not guarantee immunity to fatal accidents, since the same consequences may follow from other acts. Therefore, an increase in the frequency of decisions to quit smoking, to put on the seatbelt, or to refrain from drinking before driving, can only be expected to bring about a commensurate reduction in morbidity and mortality rates if all other relevant factors remain the same. Stated another way, this condition stipulates that people who decide to comply with the recommended safety or health behaviours do not, at the same time, adopt other behaviours that are associated with an increase in the likelihood of morbidity or mortality. One of the main contentions of this book is that this condition may not hold true and that people do change other behaviours in addition to those targeted by interventions.
An increase in seatbelt wearing does not imply fewer fatal or injury accidents. Quitting cigarette smoking does not imply an increase in the quitters' lifespan. A reduction in the amount of drinking and driving does not imply a reduction in the overall number or severity of accidents. With respect to AIDS, it has been suggested that the increased use of condoms, as propagated by mass communication messages and other means, can only be expected to slow the spread of the disease to the extent that the objective reduction in the danger of contamination exceeds the subjective reduction.[27] In other words, if mass communications would cause people to feel that "safes" are safer than they are, their sexual behaviour may change to the effect that they will be more endangered. A doctoral psychology student at my university (not one of my students) investigated the effect of televised AIDS messages produced by the Ontario Ministry of Health upon the intent to use condoms and the intent to avoid casual sexual partners. Several weeks after viewing the messages, the individuals who reported that they were more inclined to use condoms also reliably indicated that they were less inclined to avoid casual sexual partners. Not surprisingly, these results were explainable in terms of risk homeostasis theory.[28]
Regardless of whether or not this is the correct interpretation, we seem to be dealing here with a case in which the danger of a lulling effect is far from imaginary. It seems that many public health authorities feel that condom use is about the only feasible preventive measure against the threat of AIDS and, therefore, propagate it as "safe sex". Is this responsible behaviour?
While the choice of yardstick is one factor that may cause uncertainty regarding message effect, another is the manner in which an attempt is made to establish the connection between message characteristics and "effect" variables. To illustrate the interpretation ambiguities which are associated with studies that are merely correlational in nature, suppose a researcher finds in a sample of respondents a positive association between the frequency of watching alcohol advertisements and the frequency and/or amount of drinking. There are, in principle, three possible interpretations:
The direction of causality is generally obvious in investigations that take the form of a controlled experiment. It is for this reason that evidence from experiments conducted by the Addiction Research Foundation in Toronto is considerably more compelling. These experiments involved exposing audiences to alcohol advertisements or drinking scenes on television, and monitoring their drinking. None of the experiments showed convincing evidence that exposure to alcohol advertisements increased the amount of drinking by the subjects.[29] However, the contrived nature of these experiments should be acknowledged and the nagging question remains as to what extent the findings may be generalized to real life.
The body of literature on the effect of mass media communications in the area of health and safety clearly shows that these messages can have a significant effect upon people's knowledge, attitudes and behaviours. What the literature also shows is that some messages have a much greater effect than others, and the characteristics of the more effective communications have been identified.[30,31]
An obvious condition for persuasion to occur is that the position promoted by the message is different from the one already held by the recipient. Less obvious is how much change should be advocated for a message to have maximal effect. Counter to some intuitive expectations, research has shown that it is not true that the greater the change being advocated, the greater the change that will occur. Instead, what has been found is that advocated change should not exceed the "latitude of acceptability".
If, however, the advocated change does exceed the latitude of acceptability, no actual change will occur or, worse still, the message may produce a boomerang effect. The recipients do change their position, but in a direction opposite to what is being advocated in the message.
An interesting boomerang effect occurred as the result of a 60-minute televised safety programme regarding drinking and driving. A sample of 600 families in four US cities were interviewed on their perceptions of accident causation before and after the airing of the programme. Counter to expectation and intent, the programme appeared to decrease the anxiety level about travelling on holiday weekends. This was judged to be due to the fact that the programme focused on one source of danger only, namely, drinks. There was a 20% upward shift in the number of interviewees who felt that the increased risk did not apply to them, since the programme's narrow approach allow viewers to attribute the likelihood of accidents to heavy drinking, not their own behaviour.[33]
The behavioural importance of the concept of latitude of acceptability is neatly illustrated in an experiment conducted by the Austrian Road Safety Council. On a section of a six-lane highway just outside Vienna, an effort was made to reduce drivers' speed, which averaged well over 100 km/h before the experiment. In one period, a traffic sign (also a form of mass communication) was installed that indicated a speed limit of 80 km/h. In another period, the posted limit was 100 km/hreffect than the 80 km/h sign.[34]
The use of a paternalistic or lecturing approach in the presentation of the message is generally to be avoided because most people, when attending to radio or television, are usually not in the mood to be taught, but to be entertained. For greater effectiveness, messages should nonetheless be marked by concrete instructiveness. General slogans such as "Safety first", "Live a healthy life", "Safety is no accident", or "Alcohol kills slowly" (that's OK, who's in a rush?), cannot be expected to have much effect. What action is one supposed to take? The message should clearly spell out what specific behaviour is being advocated. Something like "Had a few drinks? Get a ride!" might be preferable.
Messages tend to be more effective if designed such that they are perceived as personally relevant by the recipient. They should not blame "the other guy" as the main culprit of the problem, as in the boomerang case above. Messages should enhance the processes of modelling and imitation, and the target behaviour should be displayed in the message. In order to make the audience react, the message should create a motivational state in the recipient. Motivating appeals pertain to conditions that individuals either try to achieve (romance, belongingness, prestige, self-actualization) or try to avoid (horror, pain, death, grief or ridicule).
Research on the differential effectiveness of various motivation-oriented appeals promoting the same behaviour has not been very extensive and the findings are not very clear, with two major exceptions: the use of humour has not been shown to be effective in the promotion of safe behaviour; and the use of fear is beset with major difficulties.
Strong fear appeals are aversive and lead to what is called "defensive avoidance": the recipients turn their attention away from the message, either at the very instant of exposure or by refusing to think about the message later. Mild or intermediate fear appeals can be useful, provided that the audience has an immediate opportunity to take the advocated action so that the induced state of anxiety will effectively be reduced.
Two factors are of major importance in the choice of channel or channels for communication. The choice should take account of the exposure rate in the audience as a whole or in population subgroups that one wishes to reach. When the purpose of the message is to bring about a direct effect in the form of a behaviour change, the principle of immediacy applies. Messages are more likely to have such an effect to the extent that they reach the audience at a moment, and in a situation, in which the advocated behaviour can actually be displayed by the recipients. So, television is a low-immediacy channel for messages recommending seatbelt use or driving with headlights on when it's raining. It's a high-immediacy channel for alcohol advertisements, as is countertop advertising for cigarettes in stores.
Exposure to the message is usually under the control of people themselves and will reflect their attitudes and behavioural dispositions. As the saying goes: "Volvo owners read Volvo advertisements." This does not mean that the more Volvo advertisements are printed or aired, the more Volvos will be purchased. Nor does it mean that if more of these advertisements, and those of competing manufacturers, are distributed, more cars of all makes will be sold. What it does mean is that people who have purchased a Volvo are more likely to read Volvo advertisements than the advertisements for other car makes.
The interpretation put forward by social psychologists for the fact that people selectively expose themselves to advertisements that agree with choices already made is that they seek after-the-choice justification for their decision. Voters expose themselves more to political messages that agree with their own views. Cigarette smokers pay more attention to cigarette advertisements than do non-smokers.[35]
This is why the mere observation of a positive correlation between exposure to a given type of message and a particular behaviour does not allow the conclusion that the exposure in question causes that behaviour or enhances its frequency. Yet, such claims are made, and they are made frequently.[36]
There appears to be general consensus among mass communication researchers that mass media messages, if effective at all, usually have a direct effect only on a relatively small number of recipients. These people may then pass on their behaviour change to other individuals in face-to-face interaction. The opinion leaders--the early adopters--in this two-step or multi-step flow of mass communication can be found in all layers of the population. Opinion leadership is not a general characteristic of individuals, but depends upon the particular behaviour domain involved: fashion, diet, political choice, etc. Individuals who are opinion leaders in one of these domains may well be opinion followers (late adopters) in other domains.[37]
The face-to-face part of this multi-step process is enhanced when the behaviour is conspicuously displayed for all to see. This is why mass communications for safety tend to have greater effect when the advocated change concerns a conspicuous target behaviour, for instance, driving with low-beam headlights on under conditions of reduced visibility. Being a smoker or drinker is a more visible characteristic of a person than being a non-smoker or a drinker on rare occasions only. Thus, for the purpose of enhancing the effectiveness of pro-health or pro-safety messages, it may be helpful if the personal influence link in the mass communication chain is strengthened. The early adopters may be deliberately encouraged to clearly identify themselves as non-smokers or moderate drinkers, just as people often spontaneously wear buttons or other signs in order to be perceived by others as supporters or opponents of some cause. In the area of road safety promotion, this tactic has been used deliberately and with success to increase the not-so-very conspicuous practice of seatbelt wearing. In comparison with a control area, voluntary seatbelt wearing doubled in a part of France where seatbelt users were encouraged to put a bumper sticker on their car that said; "I wear my seatbelt. How about you?"[38]
As has been demonstrated by numerous studies, mass media messages for safety and health, if well designed, can have a considerable influence on the general public's knowledge, attitudes and observable behaviours. The fact remains, however, that these messages must compete for attention with numerous other media messages and may, therefore, attract relatively little attention.
Accident reports in the daily press, on the other hand, have been found to be among the most frequently read material of all newspaper content, but they are usually not very educational in the sense of communicating to the readers how they themselves can avoid accidents. For this reason, we collaborated with a local Ontario newspaper (daily circulation approximately 35,000) to develop a prototype of a more educational accident reporting style. Over a period of eight weeks,[39] accidents were reported with more detail about the causal chain of events. They were described in the human context of antecedents and aftermath, and weekly accounts of local accident statistics were given--including date, location and severity.
A crucial feature of this "educational accident journalism" was the fact that information from accident research and documentation in general was included in the accounts of specific accidents. For example: the role of alcohol, driver age and sex, and seatbelts.
Social surveys, conducted before and after the eight-week application of the modified accident reporting style, took the form of telephone interviews with a total of 1200 licensed drivers in the experimental and in a control community. Significant changes were observed in people's opinions of the way accidents were reported, in their perceptions of the size of the accident problem, and in the importance of the accident problem relative to other issues of social concern.
A special feature of this experiment was that the reporting on the occurrence of specific accident events was used for the dissemination of general knowledge on accident causation and prevention. In that respect it was different from other efforts at educational journalism. [40,41]
As noted earlier, only a relatively small proportion of mass communication programmes for health and safety has ever been subjected to effectiveness evaluation. Moreover, the programmes that have been evaluated have often been evaluated in a manner that lacks scientific rigour. Both facts have repeatedly been lamented in the past. Even in a problem area as threatening and topical as AIDS, the many campaigns conducted in many countries have rarely been investigated on their effects. Why might this be so? One researcher commented that systematic evaluations of media health education materials are the exception rather than the rule. He attributes this to a lack of funds and to "pressures on government to deliver their campaigns to the voting public with minimal delay".[42]
From this explanation, new questions arise: why the lack of funds, and why the rush? A possible answer is the following: "For the people involved in these programmes, it becomes important that the public body be seen as responding to people's concern and that what the public body does be seen as successful. . . . By the time estimation of programme effect becomes possible, the public body has already developed a large stake in its success. Under these circumstances, why would the stewards of public bodies wish to find out what effect their programme has had? Nobody is attracted by the possibility of political, institutional, professional or personal embarrassment."[43]
A Norwegian researcher has commented that there are some forces in society resisting evaluation of traffic safety education, and that insistence upon the need to assess actual effects may be viewed as "disloyalty to the established institutions of society".[44] The true effects remain largely shrouded, that is, shrouded inside the emperor's new clothes.
Effectiveness evaluation of mass media programmes is rare. Effectiveness evaluation of mass media programmes against the criterion of accident reduction--rather than attitude or behaviour change--is rarer still. One such campaign (in fact, the only one I know of) that seemed successful in this respect was the propagation of the "Green Cross Code", a repertoire aimed at teaching children seven years and older to judge when it is safe to cross a street:
- First find a safe place to cross, then stop.
- Stand on the sidewalk near the curb.
- Look all round for traffic and listen.
- If traffic is coming let it pass, look all round again.
- When there is no traffic near, walk straight across the road.
- Keep looking and listening for traffic while you cross.
Some seven million brochures explaining the code and providing a tear-off slip were distributed in Great Britain. These were to be completed by parents certifying that their child had demonstrated understanding of the code by guiding their parents over the road three times. The total media coverage included television, posters and announcements in movie theatres. It was calculated that the average member of the audience had five opportunities to see the messages on television, and fourteen in the press.
In addition to roadside observation of children's behaviour, casualty rates projected from long-term trends for this period (in 1971) were compared with actual accident rates. With the application of conservative criteria, an eleven percent drop in the casualty rate of children between five and nine years old was found during the three months in which the Code was heavily publicized. It was calculated also that the monetary savings in medical care outweighed the programme expenditures.
These results look surprisingly good, but there are some puzzling features. According to some researchers, pre- and primary schoolchildren simply do not have the perceptual, motivational and judgemental maturity to learn to meet the demands of modern traffic. Adults, parents, drivers, or city planners, should take responsibility for children's safety. The apparent contradiction would be resolved if the success of the Green Cross Code was indeed largely mediated by active participation by the parents rather than by a direct mass media influence on the children. It was found that, in the programme period, there was not only a reduction in casualties in the target age group, but also among adults. In one geographical area, the frequency of television broadcasting of the Code was doubled, but this produced no discernible result.[45]
Astounding as it may seem, it's fair to conclude this chapter by observing that convincing proof of the effectiveness of education--be it face-to-face or by the mass media--as a means to reduce accidents and improve lifestyle-dependent health still remains to be delivered. Large amounts of money are involved. The saying has it that the world wants to be deceived. People are willing to pay for being lulled into illusions. But at this price?
[TABLE OF CONTENTS] [PREVIOUS CHAPTER] [NEXT CHAPTER]